rdblue commented on a change in pull request #2472: URL: https://github.com/apache/iceberg/pull/2472#discussion_r615322440
########## File path: spark3/src/main/java/org/apache/iceberg/spark/source/SparkFilesScan.java ########## @@ -0,0 +1,123 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, + * software distributed under the License is distributed on an + * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY + * KIND, either express or implied. See the License for the + * specific language governing permissions and limitations + * under the License. + */ + +package org.apache.iceberg.spark.source; + +import java.util.List; +import java.util.Map; +import java.util.Objects; +import org.apache.iceberg.CombinedScanTask; +import org.apache.iceberg.FileScanTask; +import org.apache.iceberg.Table; +import org.apache.iceberg.encryption.EncryptionManager; +import org.apache.iceberg.io.CloseableIterable; +import org.apache.iceberg.io.FileIO; +import org.apache.iceberg.relocated.com.google.common.base.Preconditions; +import org.apache.iceberg.relocated.com.google.common.collect.ImmutableList; +import org.apache.iceberg.relocated.com.google.common.collect.Lists; +import org.apache.iceberg.spark.FileScanTaskSetManager; +import org.apache.iceberg.spark.Spark3Util; +import org.apache.iceberg.spark.SparkReadOptions; +import org.apache.iceberg.util.PropertyUtil; +import org.apache.iceberg.util.TableScanUtil; +import org.apache.spark.broadcast.Broadcast; +import org.apache.spark.sql.util.CaseInsensitiveStringMap; + +import static org.apache.iceberg.TableProperties.SPLIT_LOOKBACK; +import static org.apache.iceberg.TableProperties.SPLIT_LOOKBACK_DEFAULT; +import static org.apache.iceberg.TableProperties.SPLIT_OPEN_FILE_COST; +import static org.apache.iceberg.TableProperties.SPLIT_OPEN_FILE_COST_DEFAULT; +import static org.apache.iceberg.TableProperties.SPLIT_SIZE; +import static org.apache.iceberg.TableProperties.SPLIT_SIZE_DEFAULT; + +class SparkFilesScan extends SparkBatchScan { + private final String taskSetID; + private final Long splitSize; + private final Integer splitLookback; + private final Long splitOpenFileCost; + + private List<CombinedScanTask> tasks = null; // lazy cache of tasks + + SparkFilesScan(Table table, Broadcast<FileIO> io, Broadcast<EncryptionManager> encryption, + boolean caseSensitive, CaseInsensitiveStringMap options) { + super(table, io, encryption, caseSensitive, table.schema(), ImmutableList.of(), options); + + this.taskSetID = options.get(SparkReadOptions.FILE_SCAN_TASK_SET_ID); + + Map<String, String> props = table.properties(); + + long tableSplitSize = PropertyUtil.propertyAsLong(props, SPLIT_SIZE, SPLIT_SIZE_DEFAULT); Review comment: I think there's still some value in a file open cost that isn't 0. The same thing we are trying to avoid in normal scans could happen: over-packing data files into a single task causing it to take forever. But it does mean that we end up with results that are possibly not idempotent. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
